Brains, Sex, and Machine Learning

Channel:Google TechTalks · Length:1:00:19 · Published:2012-08-17 · Views:151842

Watch on YouTube

Summary

Geoffrey Hinton's Google Tech Talk linking machine learning to biology, explaining dropout as a form of model averaging and why sexual reproduction and stochastic spiking neurons may prevent overfitting.

Why HN recommends it

Recommended as a simple, elegant explanation of dropout and its connection to how the brain works.

If you want a great overview of the dropout technique, watch this tech talk by Hinton — HN, 2013

Machine learning & deep learning · Neuroscience & the brain

Subtitles (excerpt)

Welcome. I'm I'm really honored to have to be able to introduce our speaker today. Jeff Hinton is from the University of Toronto. He graduated from Cambridge in 1970 with a Bachelor of Arts in experimental psychology and from Edinburgh in 1978 with a PhD in artificial intelligence. We barbarians south of the Canadian border don't take much stock in pedigree, but as a computer scientist I was couldn't help but notice that Jeff is the great great-grandson of the logician George Boole. Very cool. Among his many achievements Jeff co-invented Boltzmann machines along with Terry Sejnowski and introduced backpropagation for training multi-layer neural networks with Paul Werbos, David Rumelhart, and Ron Williams. He's the founding director of the Gatsby Computational Neuroscience Unit at University College London, which is one of the premier places to study computational neuroscience in the world. His students comprise a stellar list of computer scientists in machine learning and computer vision. They include Rick Szeliski, Richard Zemel, Radford Neal, Carl Rasmussen, Brendan Frey, and Yee Whye Teh, all research scientists well-known by many of us. He was elected to the to the Royal Society in 1998, received the IJCAI

On Hacker News

I would highly recommend this video [VIDEO] from Geoffrey Hinton in 2012 that I think is related to this research. — eveningcoffee, 2015-11
It seems the relationship between the brain and deep learning has evolved in such a way that the later can help with insights into how the former works. In this regard, I thought I would mention the extraordinary simple and elegant talk by G. Hinton last summer: [VIDEO] It starts from a simple and clever improvement to an existing deep learning method and ends up with beautiful (and simple!) insi… — pilooch, 2013-05
Here is a GREAT talk by Geoffrey Hinton (the Prof running said lab) [VIDEO] where he explains the method. Unfortunately, even though it was posted three times to HN [LINK] it never made the fron page. Here is the my summary and comment: " Great talk. I don't know much about artificial neural networks (ANN) and even less about natural ones, but I have the feeling that I learnt a lot from this vide… — FrojoS, 2012-11
As far as I know there is a decent amount of information against the fact that brains are quantum computers. On the other hand there is a talk by Hinton on youtube [1] that sheds interesting light on some variants of deep learning and the way the brain uses (classical) noise. [1] [VIDEO] — carlob, 2014-02
Using dropout with deep neural networks is an (extremely cheap) way to gain many of the benefits of using an ensemble. If you want a great overview of the dropout technique, watch this tech talk by Hinton: [VIDEO] Also, the recent maxout algorithm (from Montreal group, authors of Theano) that got state of the art results on several datasets is essentially just an algorithm designed to do particul… — lightcatcher, 2013-05

Related videos